让机器读懂论文里没写清楚的隐性知识，自动复现研究代码。
What Papers Don't Tell You: Recovering Tacit Knowledge for Automated Paper Reproduction
- 构建图神经网络代理，分三步挖掘论文中隐藏的关联、执行和群体知识。
- 在跨领域40篇论文上复现准确率提升24.68%，差距缩小至10.04%。
- 适合想自动化复现实验或理解论文深层逻辑的研究者。
自动化论文复现——从学术论文生成可运行代码——的瓶颈不在于信息检索，而在于论文不可避免地省略了隐性知识。我们将其形式化为逐步恢复三类隐性知识：关系型、身体型与集体型知识，并提出 method，一种基于图的智能体框架，针对每类知识设计专用机制：节点级关系感知聚合通过分析目标论文与其引用文献间实现单元的复用与改造关系，恢复关系型知识；执行反馈优化通过运行时信号驱动的迭代调试，恢复身体型知识；图级知识归纳则从具有相似实现模式的论文集群中提炼集体知识。在涵盖3个领域、10项任务、40篇近期论文的扩展 ReproduceBench 上， method 平均性能差距仅为10.04%（相比官方实现），相较最强基线提升24.68%。代码将在论文接受后公开，最终版本将提供仓库链接。
原文摘要 · Abstract (English)
Automated paper reproduction -- generating executable code from academic papers -- is bottlenecked not by information retrieval but by the tacit knowledge that papers inevitably leave implicit. We formalize this challenge as the progressive recovery of three types of tacit knowledge -- relational, somatic, and collective -- and propose \method, a graph-based agent framework with a dedicated mechanism for each: node-level relation-aware aggregation recovers relational knowledge by analyzing implementation-unit-level reuse and adaptation relationships between the target paper and its citation neighbors; execution-feedback refinement recovers somatic knowledge through iterative debugging driven by runtime signals; and graph-level knowledge induction distills collective knowledge from clusters of papers sharing similar implementations. On an extended ReproduceBench spanning 3 domains, 10 tasks, and 40 recent papers, \method{} achieves an average performance gap of 10.04\% against official implementations, improving over the strongest baseline by 24.68\%. The code will be publicly released upon acceptance; the repository link will be provided in the final version.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。